feat: publish the catalog as a page built from the repository - #51
Merged
Conversation
This was referenced Jul 30, 2026
HuggeK
marked this pull request as ready for review
July 30, 2026 10:34
The drivers were only browsable by reading 80 manifests. This adds a catalog page that anyone can search by manufacturer or model number, built by tools/generate_site.py from the manifests and the Lua sources themselves and republished on every push to main. No rendered copy is checked in, so the page cannot fall behind what a host installs. The hard part is not the list, it is the evidence. A catalog that shows 80 logos implies 80 working integrations. This one carries the claim the driver actually makes: 5 drivers say they have run on physical hardware, 28 say in their own verification_notes that they have not, and each driver's per-target conformance and HIL state are shown beside it. tests/test_site.py fails the build when a status is rendered as a stronger claim than the source makes, and when a manifest, version or tested model is missing from the page. Titles and summaries come from the drivers too: 38 carry a DRIVER table with a description, the other 42 describe themselves in a header comment that is quoted verbatim rather than paraphrased into a second copy that could drift. Signed-off-by: Hugo Karlsson <48095810+HuggeK@users.noreply.github.com> Co-authored-by: HuggeK <48095810+HuggeK@users.noreply.github.com>
lua_string_list() read `key = { "a", "b" }` fields for tested_models and
capabilities. Both were dropped before the first commit — the manifests
already carry the tested models, and capabilities restate the DER list —
so the helper was left with no callers.
Signed-off-by: Hugo Karlsson <48095810+HuggeK@users.noreply.github.com>
Co-authored-by: HuggeK <48095810+HuggeK@users.noreply.github.com>
Rebasing onto main carried this entry along inside the Added group it was written in, next to BLUEPRINT.lua, which now reads as though the catalog shipped in that batch. [Unreleased] prepends a group per merged change, so it belongs at the top. Signed-off-by: Hugo Karlsson <48095810+HuggeK@users.noreply.github.com> Co-authored-by: HuggeK <48095810+HuggeK@users.noreply.github.com>
HuggeK
force-pushed
the
driver-catalog-page
branch
from
July 30, 2026 10:37
7c19d46 to
6f8eab7
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The drivers were only browsable by reading 80 manifests. This publishes a catalog page anyone can search by manufacturer or model number, at https://srcfl.github.io/device-drivers/ once this merges.
tools/generate_site.pybuilds it frommanifests/,drivers/lua/,support-status.jsonanddriver-history.json, and.github/workflows/pages.ymlrepublishes it on every push tomain. No rendered copy is checked in —site/is gitignored — so the page cannot fall behind what a host installs.What it shows
SH10RTfindssungrow.min_driver_version, the per-target conformance and HIL state, the source sha256, and links to the Lua source and manifest.drivers.jsonbeside the page, for anyone who wants the catalog as data.The evidence is the point
A catalog that shows 80 logos implies 80 working integrations. This one carries the claim each driver actually makes. 5 drivers say they have run on physical hardware; 28 say in their own
verification_notesthat they have not, most of them "ported from a reference implementation". The page rendersproductionandtestedas confirmed,betaandalphaas partial, andexperimentalas not verified on hardware — never as a single green tick.tests/test_site.pyfails the build if any driver's status is rendered as a stronger claim than the source makes, and if a manifest, version, tier, control flag or tested model goes missing from the page.Where the words come from
38 drivers carry a
DRIVER = { ... }table with a description, homepage and verification record. The other 42 describe themselves in a header comment, which is quoted verbatim under "From the driver source" rather than paraphrased into a second copy that could drift from the code. Nothing on the page is typed by hand.GitHub Pages setup
Pages is already enabled on this repository with
build_type: workflow, and thegithub-pagesenvironment exists, so no settings change is needed — merging this fires the first deploy to https://srcfl.github.io/device-drivers/.actions/configure-pagesstill runs withenablement: true. That is a no-op while Pages is on and correctly configured; it is kept so the workflow re-enables itself rather than failing if the setting is ever reset.Checked
pytest tests/test_site.py— 8 tests pass../lua55, whichmake checkbuilds, and bash path handling); none touch these files.tools/check_public_boundary.pypasses.<noscript>, so it works without JavaScript.🤖 Generated with Claude Code